AI module CI/CD

In the platform, we use Jenkins for implementing the CI/CD (Continuous Integration / Continuous Development) pipeline. This pipeline automatically performs a series of actions for you each time you commit a change in your code. This ensures that all the information and builds across the project are always up-to-date with your code.

A typical pipeline is divided in two parts:

• a model quality assurance section, using tox, where we do:

  • code style analysis using flake8,

  • unit testing using pytest,

  • security scanners using bandit,

  • metadata validation,

• a platform components update, where we:

  • build the Docker image to the platform Registry and mirror to DockerHub,

  • update the Marketplace and additional platform services,

  • archive the code to Zenodo if a code release is made,

  • regenerate the provenance chain of that module,